bepicolombo electric propulsion thruster and high power

The 33rd International Electric Propulsion Conference, The George Washington University, USA

October 6 – 10, 2013

1

BepiColombo Electric Propulsion Thruster and High Power 1

Electronics Coupling Test Performances 2

IEPC-2013-133 3

4

Presented at the 33rd International Electric Propulsion Conference, 5

The George Washington University • Washington, D.C. • USA 6

October 6 – 10, 2013 7

8

Stephen D Clark1 and Mark S Hutchins.

2 9

QinetiQ, Cody Technology Park, Farnborough,GU14 0LX, UK 10

Ismat Rudwan3 and Neil C Wallace

4 11

Mars-Space Ltd, Southampton, Hampshire, SO14 5FE, UK 12

Javier Palencia5 13

EADS Astrium CRISA, C/. Tores Quevedo, 9(P.T.M.), 28760 Tres Cantos, Madrid, SPAIN 14

and 15

Howard Gray6 16

Astrium, Anchorage Road, Portsmouth, Hampshire PO3 5PU, UK 17

Abstract: This paper describes the equipment ‘building blocks’ (T6 thruster, power 18

supply and control unit and flow controller) that make up the two systems HPEPS and SEPS 19

that are in development and qualification for application on commercial telecommunication 20

satellites and ESA’s BepiColombo mission to Mercury, respectively. The paper focuses on 21

the SEPS Coupling Test, the results from which show the compatibility of these equipment 22

‘building blocks’ within the SEPS and the required system performance, a major 23

development milestone for SEPS towards successfully delivering the BepiColombo mission 24

to Mercury. 25

I. Introduction 26

Electric propulsion systems are being developed by QinetiQ to 27

meet two current applications. One is the High Power Electric 28

Propulsion System (HPEPS) for station keeping, orbit-topping and 29

end-of-life de-orbit for geostationary orbit (GEO) 30

telecommunications satellites. The second application is the ESA 31

science mission to the planet Mercury, BepiColombo1, where the 32

system will provide the impulse necessary during the inter-planetary 33

cruise. The BepiColombo composite spacecraft is illustrated in Fig. 34

1. 35

1 AIV Technical Lead, Space UK, [email protected]

2 HPEPS Systems Engineer, Space UK, [email protected]

3 Senior Research Engineer, Space, [email protected]

4 Chief Engineer, Space, [email protected]

5 PPU Responsible Engineer, Crisa, [email protected]

6 SEPS Responsible Engineer, Astrium, [email protected]

Figure 1. Artist’s impression of the

BepiColombo composite

Image courtesy ESA

mailto:[email protected]

mailto:[email protected]


October 6 – 10, 2013

2

QinetiQ has completed a successful Technology Demonstration Activity (TDA)2;3

of a T6 thruster to retire the 36

high risks associated with Europe’s first use of electric propulsion for a major science mission. This activity 37

investigated new requirements such as sustained, stable operation over long thrust arcs, the simultaneous operation 38

of two thrusters at high power in close proximity, low thrust interruption rates and the harsh thermal environment 39

approaching the Sun. 40

The Solar Electric Propulsion System (SEPS) for BepiColombo was awarded to QinetiQ in 2009 following an 41

open competition. SEPS is based on the HPEPS and benefits from its wide operating range, robust design and its 42

ability to be easily adaptable to a wide range of platforms and applications. 43

This paper briefly describes the common building blocks between HPEPS and SEPS and the respective 44

architectures for each application. The paper outlines the overall SEPS Assembly Integration and Verification (AIV) 45

programme before presenting in detail the Coupling Testing that has been performed to demonstrate the 46

compatibility of the components of the system: thruster, power supply and control unit and the flow control unit and 47

verify the performance of the coupled SEPS hardware. 48

II. Common building blocks and system architecture 49

A. Common Building Blocks 50 The QinetiQ electric propulsion systems are composed of the T6 thruster, a power supply and control unit (PPU 51

– developed by EADS Astrium Crisa), Xenon flow control units (FCU – developed by Moog-Bradford Engineering) 52

and associated harnesses. These hardware elements are common to both HPEPS and SEPS. In addition the 53

consortium members are the same as the successful GOCE programme4,5

and build on the heritage and experience 54

developed during this programme. 55

The resultant QinetiQ electric propulsion system is capable of achieving a wide range of operational 56

requirements, from low to high power applications whether constant thrust level, multiple thrust level or variable 57

thrust level operation. 58

Thrust vectoring is achieved in both systems via pointing mechanisms but due to the very different pointing 59

requirements these mechanisms are not common to HPEPS and SEPS. The SEPS system boundary excludes the 60

pointing mechanism and consequently is not covered in this paper. For HPEPS a pointing mechanism is being 61

developed by RUAG Space Austria (referred to as the electric propulsion pointing mechanism or EPPM) – this is 62

further described below along with the common building blocks: T6, PPU and FCU. 63

64

1. QinetiQ T6 Thruster 65

The T6 thruster was first designed in 1995 to meet 66

the needs of science and telecommunication missions in 67

the 21st century. The thruster was developed using 68

experience and scaling laws identified during the early 69

stages of the QinetiQ T5 development. Since 1995, the 70

T6 design has undergone extensive development to 71

optimise performance and lifetime and make effective 72

use of advances in materials for insulators, grids and the 73

ionisation chamber. All of this has been achieved and 74

demonstrated within the BepiColombo Technology 75

Demonstration Activity (TDA), the Grid Technology 76

Research Programme (TRP) and the HPEPS 77

development and qualification programmes, sponsored 78

by ESA. 79

The T6 thruster is shown schematically in Fig. 2. It is 80

of conventional Kaufman configuration6, with a direct 81

current (DC) discharge between a hollow cathode and a 82

cylindrical anode used to ionise the propellant gas. The 83

efficiency of this plasma production process is enhanced 84

by the application of a magnetic field within the 85

discharge chamber. 86

A 22 cm diameter grid system, forming the exit to the 87

discharge chamber, extracts and accelerates the ions, to 88

ANODE

SOLENOID

KEEPER ACCELERATOR GRID

OUTER POLE

BAFFLE

PROPELLANT

ISOLATOR INNER POLE

DISCHARGE CHAMBER

EARTHED SCREEN

PROPELLANT

INSULATOR CATHODE

NEUTRALISER ASSEMBLY

ISOLATOR

HOLLOW CATHODE

CATHODE SUPPORT

SCREEN GRID

INSULATORS

FERROMAGNETIC CIRCUIT

MAGNETIC FIELD LINE

MAIN FLOW DISTRIBUTOR

Figure 2. Schematic of T6 thruster


October 6 – 10, 2013

3

provide the required thrust. The velocity of the ejected ions depends only on the beam potential, whereas the thrust 89

is a function of this and the ion beam current. An external hollow cathode, referred to as the neutraliser, emits the 90

electrons necessary to neutralise the space charge of the emerging ion beam. 91

92

2. T6 Thruster Harness 93

At an early stage of the HPEPS programme it was recognised that the harness between the thruster and the PPU 94

was a critical system element and that it was imperative that the system engineering of the harness must be 95

performed concurrently with the thruster. The design is driven by the high voltage, high current and high 96

temperature operation in conjunction with the need to traverse the moving interface presented by the pointing 97

mechanism. It was also recognised that to allow the mechanism and harness integration to be performed in isolation 98

to the thruster (earlier thruster designs have always included an integral flying lead design) it was necessary to 99

introduce a connector block on the thruster. 100

To maximise heritage the harness design employs the same commercially available ePTFE (corona resistant) 101

coaxial high voltage cable technology qualified on the GOCE programme with the T5 ion thruster. 102

103

3. Power Supply and Control 104

The Power Supply and Control Unit (PPU) provides the required functionality to power and control the T6 105

thruster and the Xenon Flow Control Unit (XFCU) as commanded by the satellite’s on-board computer. The PPU 106

also provides the necessary telemetry and telecommand interfaces to control and monitor all thruster functions. 107

In order to minimise the mass and volume of the PPU its design takes advantage of the fact that a number of the 108

thruster and XFCU control functions can be 109

supplied by a common power supply without 110

impact to the performance or operation of the 111

system. 112

This results in a total of 5 power supplies 113

being required to power the T6 thruster. This is 114

illustrated in Fig. 3. 115

The 5 PPU thruster supplies are as follows: 116

117

1. Neutraliser Heater (Primary & 118

Redundant) + Neutraliser Keeper 119

supply 120

The keeper can be powered by the 121

same supply as the heater as the two 122

are not required to operate 123

simultaneously. 124

2. Accel Grid Supply 125

3. Beam Supply 126

4. Anode + Cathode Keeper supply 127

In the case of the Cathode keeper, it is 128

operated in parallel with the Anode 129

supply. 130

5. Magnet + Cathode Heater (Primary 131

& Redundant) supply 132

The Cathode heater and the solenoids share the same supply, as they are not required to operate simultaneously. 133

134

Where supply functions share a common transformer within the PPU, the changeover between supplies, is only 135

performed when the supply is in a shutdown state i.e. hot switching is eliminated. 136

In addition to the above nominal functionality, the PPU design makes use of the high current capability of the 137

Anode supply to provide a grid clearance function. This functionality means that should a conductive particle cause 138

an electrical short between the Accel and Screen grids, and in the event that this particle does not ‘self clear’, the 139

Anode supply can be switched to the Accel grid to ‘burn-away’ the contaminant. 140

-ve

Anode +ve

Accel. Grid

Neutraliser

Keeper

Power Processor Unit (excluding FCU drivers)

Solenoid

+ve

-ve

Beam Supply

Cathode

Neutraliser

Main Flow

So

len

oid

/ H

ea

ter

Su

pp

ly

Cathode

Heater

+ve

Neutraliser

Heater

-ve

He

ate

r /

Ke

ep

er

Su

pp

ly

-ve

-ve

Sp

lic

e P

late

Thruster

Spacecraft

Ground Figure 3. Thruster power supplies schematic


October 6 – 10, 2013

4

The PPU architecture is illustrated in Fig. 4. The PPU design incorporates a Faraday housing as was employed in 141

the IPCU for GOCE. In this design the high 142

voltage referenced power supplies; i.e. the 143

anode, solenoids, cathode keeper and cathode 144

heater supplies, are enclosed in a housing which 145

is floated at beam potential and the whole unit 146

is isolated from the unit chassis via 147

conventional ceramic isolation techniques. This 148

concept has 2 major advantages. The first is 149

that it virtually eliminates difficulties associated 150

with parasitic elements within the high voltage 151

(HV) referenced supplies, e.g. unexpected 152

current paths for the common mode currents are 153

contained within the enclosure. Secondly the 154

power supplies can be readily designed without 155

having to deal with any HV isolation design 156

complexities, i.e. isolation is achieved once using simple ceramic stand-offs, because the components within the 157

supplies are referenced to the HV enclosure. 158

The PPU is a single equipment in the T6 HPEPS. Comprising of 2 main functional elements, the beam supply 159

and the DANS [2off]. 160

Beam supply: The beam supply is comprised of up to 5 parallel beam supply modules. In the case where 3 beam 161

modules are required to fulfill the mission requirement a 4th beam supply module is also included for redundancy. 162

As such the beam supply is tolerant of a single Beam Supply Module failure, i.e. 4 for 3 parallel redundancy. 163

DANS: There are 2 DANS assemblies within the PPU, providing 2 for 1 parallel redundancy. The outputs of 164

each DANS are switched to either one of the north thrusters or one of the south thrusters. 165

The PPU is able to be a single equipment, with significant mass and cost savings, and provides all the necessary 166

redundancy and failure tolerance required within that equipment. This also enables the S/C level thermal control to 167

be simplified because the heat loads are always in the same place, i.e. provision does not need to be made for 168

dissipation in two locations. 169

170

4. Flow Control Unit 171

During thruster operation the Flow control unit 172

(FCU) when powered will deliver propellant at the 173

required mass flow rates to a single thruster. 174

The FCU uses a proportional valve in closed loop 175

control with a pressure transducer to control the 176

pressure at the inlet to a fixed restrictor (at constant 177

temperature). 178

The pressure based control method is a simple and 179

robust solution for the Main and Cathode flow control. 180

• It is a direct method of control based on the 181

basic parameters determining the mass 182

flowrate, i.e. pressure, temperature and the 183

physical restriction. 184

• Pressure sensing is considered to be more 185

dependable than mass flow sensing which is 186

less susceptible to temperature variations. 187

• It is preferable to minimise the variety of 188

functional components and control methods in 189

a single design. In this solution the Main Flow 190

and Cathode feeds will use the same flow 191

control components and method 192

193

A functional block diagram for the flow controller 194

is shown in Fig. 5. The FCU includes a filter at its 195

inlet to protect the system from particulate 196

Figure 4. Schematic of PPU architecture

Figure 5. Functional block diagram of the FCU


October 6 – 10, 2013

5

contamination. 197

Immediately downstream of the inlet filter there is a single mono-stable isolation valve, after which the flow path 198

divides into 3 parallel branches for the Neutraliser Feed, Cathode Feed and Main Flow Feed. 199

The Main Flow Feed and Cathode Feed branches each contain a proportional flow control valve, pressure 200

transducer and fixed restrictor. The Neutraliser Feed branch contains a mono-stable isolation valve and a fixed 201

restrictor only. The Neutraliser Feed can be operated independently of the Main Feed and Cathode Feed. 202

Each valve type, IV or FCV, is capable of achieving the necessary internal leak tightness when closed, and both 203

are normally closed devices so that in the event of a loss of power to those valves they would automatically close. 204

Any combination of valves can be opened independently or simultaneously for test, venting and purging operations 205

on ground or in orbit. The overall configuration of valves ensures that there are 2 in-series inhibits between the inlet 206

and each outlet, so that in the event that any one valve failing open or leaking excessively the propellant upstream of 207

the FCU can be isolated 208

209

5. Pointing Mechanism 210

The role of Electric Propulsion (EP) thrusters in future commercial spacecraft (S/C) is expanding from the 211

North-South Station Keeping (NSSK), which requires only limited angular pointing range, to include also Orbit 212

Topping and Momentum Dumping Manoeuvres, and possibly East-West Station Keeping and Orbit Raising as 213

options, in order to save fuel mass and therefore to increase the revenue gain (more payloads, increased lifetime, 214

smaller launchers, or a combination of these). 215

Existing mechanism designs have been able to facilitate limited thruster operation in the event of a mechanism 216

‘failure to deploy’ scenario, by stowing the thrusters at a nominal 45° pointing angle. However, the next generation 217

of large electric propulsion thrusters, in particular due to their size, require a significantly different type of pointing 218

mechanism. The ability to perform limited thruster operation in the failed deployment scenario has become an 219

excessive constraint on the system and platform design, to a point that it significantly impacts the considerable 220

advantages these thruster systems would otherwise provide. As such it is considered that the requirement to stow the 221

thrusters at a nominal 45° to no longer be practical for the next generation of EP systems for telecommunication 222

platforms. 223

The EPPM illustrated in Fig. 6 consists of a 224

mobile platform interfacing with a single T6 225

thruster. The thruster can be pointed around two 226

perpendicular axes due to a two-axis rotation 227

mechanism and dedicated drive units. The 228

EPPM can achieve up to 90° pointing angle for 229

the first axis and +15° pointing angle for the 230

second axis. The mobile plate is kept locked in 231

a specific reference position during the launch 232

phase by a Hold Down and Release 233

Mechanism. 234

The main parts of the EPPM are: 235

• Mobile Platform 236

• Two-Axes Pointing Mechanism 237

• Hold-Down and Release Mechanism 238

(HDRM). 239

• Damping system underneath the HDRM 240

• Harness and piping support for the T6 Thruster 241

242

Figure 6. Illustration of the EPPM, showing rotation axes


October 6 – 10, 2013

6

B. EP System Architectures 243 244

The primary function of the HPEPS when defined in 2004 was to provide on-station NSSK for geostationary 245

telecommunication satellites. HPEPS is also capable of performing satellite orbit top up and repositioning 246

manoeuvres e.g. to graveyard at the end of mission. 247

In July 2003 QinetiQ was awarded a contract by ESA to 248

perform the necessary pre development activities of an ion 249

propulsion system for the Large Platform Mission Project 250

(AlphaBus) in a NSSK role. 251

During the initial stages of the programme, in-depth 252

trade-off activities at system and equipment level were 253

performed from which the baseline system has been defined. 254

The system architecture is presented in Fig. 7 and comprises: 255

• 4 x T6 thrusters, each incorporating a dedicated 256

Neutraliser. 257

• 4 x Xenon Flow Control Unit (XFCU) assemblies, 258

each dedicated to a specific T6. 259

• 1 x PPU, incorporating two sets of Discharge, Accel 260

and Neutraliser supplies (DANS) a modular failure 261

tolerant beam supply. 262

• 2 x Electric Propulsion pointing mechanisms (EPPM) 263

• 4 x Splice plates 264

• Harness and pipe work. 265

The HPEPS is readily adaptable for a variety of 266

applications and mission types. Variations of the HPEPS 267

architecture are currently being proposed to provide electric 268

propulsion for the European Galileo Second Generation System and Electra, a telecommunication satellite. 269

270

There are two critical differences between the GEO 271

telecommunications and BepiColombo applications. The first 272

is that BepiColombo requires two thruster simultaneous 273

operations at a combined thrust level of 290 mN, hence two 274

PPUs are necessary. The second is that the 290mN must be 275

assured throughout the transfer and hence an additional 4th 276

BSM is required in each beam supply in each PPU. The SEPS 277

architecture is illustrated in Fig 8. 278

Critical to the system design is the redundancy and cross- 279

strapping philosophy. The selected configuration of two 280

PPUs with fully symmetric cross-strapping of T6 units to 281

PPU outputs gains maximum benefit from the symmetry of 282

the four engine configuration. This provides the most robust 283

response any system element anomaly. It also allows the 284

ability to maintain equal distribution of total impulse between 285

operational T6 units with any single unit failure in a BSM, 286

DANS, FCU or a T6. This in turn minimises the thruster life 287

time requirement. 288

The resulting design provides for an elegant AIV 289

sequence and plan. The number of units is minimised (only 290

two PPUs without auxiliary electronic boxes), mounting and 291

demounting units is straightforward with the integrated splice 292

plate and the number of operating modes for testing is 293

minimised. In addition, system mechanical and electrical symmetry mean that thermal behaviour will have fewer 294

operational patterns to verify. 295

Figure 7. Architecture for commercial

communications satellite (NSSK and orbital re-

positioning).

PS

CU

#2

PS

CU

#1

BSM#1a

BSM#1b

BSM#1c

BSM#1d

DANS#1

DANS#2

T

S

U

#

2

T

S

U

#

1

BSM#2a

BSM#2b

BSM#2c

BSM#2d

DANS#3

DANS#4

T

S

U

#

4

T

S

U

#

3

FCU #4

T6#1

FCU #1

FCU #2

FCU #3

T6#2

T6#3

T6#4

2.5 – 2.85bar

S/C

Power

and

Control

SP

SP

SP

SP

S/C

Power

and

Control

Figure 8. SEPS architecture providing guaranteed

simultaneous thruster operation and impulse sharing


October 6 – 10, 2013

7

III. SEPS AIV Program 296

The aim of the SEPS AIV program is to verify the EQM and FM SEPS hardware in a representative (to the flight 297

configuration) test set-up and verify SEPS requirements. The SEPS AIV program is divided into the following 298

important activities, and provides the opportunity for an early development test. 299

C. SEPS Coupling Test Part 1 300 The main purpose of this test campaign was to characterise the performance of coupled SEPS hardware. It also 301

demonstrates specific PPU-thruster interactions including the FCU control loops, beam-out recovery PPU automatic 302

command and control sequences, nominal/redundant harness characterisation and FCU control algorithms. The PPU 303

and FCU are mounted outside of the vacuum facility to allow attachment of key diagnostic equipment. 304

D. SEPS Coupling Test Part 2 305 Coupling Test Part 2 provides a complete representative chain, with the PPU, FCU and Thruster mounted inside 306

the vacuum chamber with representative connections for the SEPH and SEPP. Performance tests of the SEPS will be 307

conducted, and validation of key SEPS requirements will be performed during this test phase. 308

E. SEPS Endurance Test 309 The SEPS hardware (Thruster, PPU, FCU, SEPH and SEPP) will be exposed to an 8000 hour endurance test. 310

The purpose of these system-level tests is to verify if and how the other SEPS components will influence the long- 311

term performances and wear-out mechanisms of the thruster (and therefore its life-time). The Endurance test shall 312

address the following requirements: Total impulse and total firing time for each thruster/FCU assembly, thruster ion 313

optics modeling and lifetime assessments shall be performed, each thruster, FCU and PPU shall be capable of at 314

least 500 operational cycles, SEPS Autonomy and Availability, thrust vector offset and stability and thruster plume 315

beam divergence less than 25º half cone angle, at all stages of the thruster life. 316

F. SEPS EMC 317 The EMC test campaign will entail measurements of power quality, conducted and radiated emissions, and 318

radiated susceptibility of the thruster driven by the EQM PPU. The measurements require the thruster to be inside 319

an electromagnetically screened enclosure to reduce the background RF noise to minimum levels. The emissions 320

and susceptibility of the thruster can then be measured using a series of different antenna, receivers and transmitters 321

covering the frequency range of interest. All of the thruster EMC testing will be conducted with the thruster 322

operating inside an RF transparent section of the vacuum chamber, shrouded by the screened anechoic room. 323

G. SEPS Acceptance Test. 324 The FM SEPS acceptance test campaign will be a complete end-to-end SEPS test involving all of the SEPS 325

flight hardware and tested in vacuum chambers. Beam probe measurements will also be taken during this test 326

campaign to demonstrate that the thrust vector is not affected when a second thruster is in operation in dual mode. 327

For technical and programmatic reasons, this test has been split into two campaigns, each testing half of the 328

complete SEPS assembly. 329

IV. SEPS (EQM) Coupling Test 330

A. Objectives 331 The purpose of the SEPS (EQM) Coupling Test was to perform compatibility tests between the PPU, Thruster 332

and FCU as a confidence verification of the coupled SEPS hardware. The confidence checks are primarily limited to 333

electrical compatibility, FDIR checks and demonstration of beam-out tolerance/beam-out recovery. Due to the 334

configuration of the test (PPU and FCU outside of the vacuum chamber) it was possible to exploit the use of GSE 335

diagnostics to allow for interrogation of key interfaces and verification of some requirements that would not have 336

been possible with the hardware mounted inside the chamber. 337

338

The SEPS Coupling Test Part E-F provides confidence checks of the following: 339

340

Electrical compatibility of PPU, EM FCU & Thruster 341

Assessment of PPU automatic command and control sequences 342

Assessment of PPU FDIR functionality 343


October 6 – 10, 2013

8

Assessment of FCU control algorithms and Thruster/PPU coupling 344

Assessment of nominal SEPT operation at min/max operating criteria for PPU input voltage and FCU 345

inlet pressure 346

Nominal and redundant (cross-strapped) harness compatibility 347

B. Test Set-Up 348 The Coupling Test was executed with the PPU and FCU located outside of the vacuum facility, this allowed 349

attachment of key diagnostic equipment, to validate the key interface (mainly PPU to Thruster and FCU to 350

Thruster), this was achieved by attaching 351

Voltage/Current probes on the primary and 352

secondary lines of the PPU which allowed data to 353

be collected on the steady-state and transient effects 354

during operation. On the outlet of the FCU, three 355

pressure transducers were attached to the pipework 356

to allow information to be gathered on the 357

immediate downstream pressure of the FCU during 358

the dynamic and steady-state performance tests, this 359

allowed us to assess the performance of the FCU 360

flow control valves. All of the above would not 361

have been viable inside of the vacuum chamber. 362

Figure 9 shows the top level schematic of the 363

test set-up, whist Fig. 10 shows a picture taken of 364

the test that was conducted in QinetiQ’s LEEP3 365

(Large European Electric Propulsion facility No.3) 366

Test Facility earlier this year. 367

The main objective of the test from the outset was to ensure that all of the interfaces between the SEPS hardware 368

are flight representative (i.e. harness and pipework), taking in to account the electrical and fluidic feed-throughs to 369

break the vacuum vessel. With regards to the pipe interface, apart from the small volume increase due to the addition 370

of the inline pressure transducers (minimal overall 371

impact), the pipe length, diameter and wall thickness 372

were selected to exactly match the flight interface and 373

design. The harness, for obvious reasons had to break 374

the vacuum seal, and be long enough to attach to the 375

PPU (mounted externally to vacuum vessel) and route 376

around the inside of the chamber, this led to an overall 377

increase in test harness length compared to the flight 378

system. This was adequately addressed with the 379

following design enhancements (for GSE replica harness 380

only). 381

382

383

384

385

386

387

1. SEPH Test Harness Design 388

Figure 11 shows a simplified illustration of the cable cores that comprise the BepiColombo spacecraft electrical 389

harness between the PPU and SEPT splice-plate (TSP). The illustration shows the SEPT electrical circuit elements 390

that are served by the harness, and that several of those circuit elements utilise two or more cores in parallel to 391

provide adequate current carrying capacity. When it came to performing the coupling tests between the PPU and 392

SEPT, the necessary test configuration of having the SEPT in a vacuum chamber and the PPU outside the chamber 393

precluded the use of a test harness constructed to the same design as the spacecraft harness, yet there was a desire 394

that the test harness should be as representative of the spacecraft harness as possible. 395

LEEP-3 Clean Tent

Interface plate

LEEP3

H:PPU/DE/1 (Crisa)

Support Frame

Elec FT EQM/TSP/1a

EQM FCU

HCU/1P:HCU/1-PPU/TIP

(isolated)

EQM

SEPTBB SEPT

HPRS/DE/A

PSCU/DE (Crisa)

H:PPU/EQM-FT

P:H

PR

S/A

-FC

U/1

H:HPRS/DE/A-HPRS/A

H:L1/FT-EQM/TSP/1a

H: T

IPC

/3-H

CU

/1

TC Logger

TIPC/3

Elec FT

H:TIPC/3-PPU&FCU/TC

H:TIPC/3-FT/SEPT/TC

H:L1/FT/SEPT/TC/a-SEPT

H:TIPC/3-PPU/TIP/TC

6TCs

H:SEPH/EQM/BB/Flex

6TCs 6TCs

H:E

QM

/FC

U.1

HPRS/A

P:R

eg

/1-H

PR

S/A

Fluid FT

H:FMS/1-PPU/DE

6TCs

PPU EQM

Be

am

Su

pp

ly

[3 B

SM

]

H:EQM/Fix/X/PS.4a

DANS #1

EMPTY

SLOT

EQM/CSP/1

6TCs

Intention is to design a single

harness that suits both the

isolated PSCU and non-

isolated PPU requirements

Single Crisa cable for both

PSCU & PPU testing

P:L1/FCU&Brnk/A-SEPT

Facility Ground

Signal Ground

PT PTPT

Facility Monitoring System /2

Reg/1

Xe

Xe

Bronkhorst/3

Facility Monitoring System /1

Figure 9. Schematic of SEPS (EQM) Coupling Test

Figure 10. Picture of the SEPS (EQM) Coupling Test

Set-Up


October 6 – 10, 2013

9

Owing to the constraint imposed by the physical location of the PPU and SEPT with respect to the position of 396

the vacuum chamber electrical feed-through port, the 397

test harness length needed to be almost twice as long 398

(7 m) as the spacecraft harness (nominally 4 m). This 399

additional length would increase the harness electrical 400

resistance and, with the additional contact resistance 401

of the feed-through connectors, there was concern that 402

this additional voltage drop could affect the operation 403

of the Thruster. The test harness was therefore to be 404

designed with flight representative resistances – 405

particularly on the lines carrying the highest currents. 406

The test harness design that successfully fulfilled 407

the requirements is shown Fig. 12. 408

Compared to the spacecraft harness illustration 409

shown earlier, it can be seen that several of the multi- 410

cored lines have been reduced to just one. To maintain 411

current carrying capacity and minimise their 412

resistance, thicker gauge cable has been used for these 413

and other lines. 414

In order for the harness to be attached to the PPU and SEPT splice-plate terminals, the same connection layout 415

and core configuration as on the spacecraft harness design had to be utilised. 416

In order to provide confidence that the test harness design performance would be satisfactory, an analysis was 417

performed to compare the electrical impedance characteristics of the test and spacecraft harnesses. 418

The analysis of volt-drop due to resistance showed 419

that, in the anode circuit, a volt-drop of just 200 mV 420

higher with the test harness than the spacecraft harness 421

could be expected and that this increase could not 422

adversely affect the Thruster performance. Apart from 423

a small (1 mV) volt-drop increase on the Acceleration 424

grid line of the test harness, all other test harness lines 425

showed a reduction in volt-drop compared to the 426

spacecraft harness. 427

Although the Thruster is driven by DC power 428

supplies, its dynamic plasma discharge behaviour and 429

transient beam-out events mean that voltages and 430

currents have frequency/time dependencies that could 431

be affected by the reactive properties (capacitance and 432

inductance) of the harness to the extent that Thruster 433

performance is affected. Estimates indicated that the 434

test harness capacitance could be more than twice that 435

of the spacecraft harness and that the inductance maybe up to five times greater for some lines. However, other 436

reactive components in the PPU/Thruster circuits, such as power supply output capacitors, have values that are one 437

or more orders of magnitude greater that the total harness values. This meant that, despite relatively large differences 438

in capacitance and inductance between the test and spacecraft harnesses the differences were unlikely to affect 439

Thruster performance. 440

C. Performance 441 1. SEPS Operation Modes and FPGA Functions 442

This Section presents an overview of the manner in which a SEPS functional chain is controlled. A SEPS 443

functional chain is defined as a thruster, its associated FCU, any electronics which is dedicated to that thruster, and 444

all interconnecting harness and pipe-work. 445

As the primary propulsion system of the spacecraft during interplanetary flight, the SEPS takes demands from 446

the spacecraft and delivers the required thrust. Firmware resident within a Field Programmable Gate Array (FPGA) 447

in each DANS, controls the output of various power supplies and FCU, and allows the system to be throttled 448

between any thrust levels within the operational range. 449

Figure 11. Representation of SEPS Harness Wiring

Figure 12. Representation of SEPS ‘Test’ Harness

Wiring


October 6 – 10, 2013

10

The FPGA interfaces to all of the power supplies and FCU drivers (i.e. FCU controllers) within each DANS and 450

also commands the Beam supply modules. The FPGA is programmed to perform a number of tasks which include 451

the following. 452

Communications interface to the spacecraft On-board Computer (OBC). 453

Thruster and PPU safety interlocks. 454

Command and control of all the power supplies and FCU drivers. 455

Sequences for the automatic operation of the associated Thrusters and FCUs. 456

Provision of SEPS FDIR 457

Provision of all Telemetry 458

Each functional chain of the SEPS can be commanded either manually or automatically. In Manual mode the 459

FPGA can be commanded via direct external commands to control all power supplies and FCUs as instructed, with 460

the exception of commands prohibited by the safety interlocks. In automatic mode a SEPS functional chain is 461

commanded using sequences internal to the FPGA designed to perform the following functions. 462

Neutraliser start up 463

Main Discharge Cathode start-up 464

Discharge only operation 465

Thrust control 466

Grid short clearance 467

Figure 13 presents the top level State diagram for a single DANS and shows the relationship between DANS 468

states. The green state is the OFF state where the system is completely shut down. The blue states are those that 469

allow system setup and FCU activities, and the purple states are those that allow other power functions. 470

State transitions may be induced by four possible sources: On/Off electrical signal (labeled as HP in the figure) 471

from the spacecraft, commands issued by the OBC (labeled as S/C in the figure), FDIR internal to the DANS and by 472

normal progression through a control sequence (labeled as Auto in the figure). Each state transition is numbered and 473

a key identifies which source(s) may cause that particular state change. 474

In manual operation each DANS will allow the 475

operation of PPU and FCU functions as demanded 476

directly by the OBC or ground operator. Thruster out- 477

gassing and FCU venting and purging are conducted 478

under manual control (in MANUAL state). 479

In Automatic Mode the operations are designed to 480

provide hands free starting of the thrusters and maintain 481

thrust control. Each DANS is designed to automatically 482

sequence the power supplies and FCU drivers to allow 483

the following automatic sequences to be performed. 484

starting of the cathodes, 485

maintaining the discharges, 486

controlling the thrust (with beam on, also 487

includes automatic beam-out recovery), 488

automatic short clearance 489

During THRUST CONTROL each DANS allows 490

updated operating parameters from the OBC to be 491

adopted to allow the thrust level from any operating Thruster to be varied without interruption in operation. 492

Transition between Thruster DISCHARGE and THRUST CONTROL shall be commanded via the OBC. 493

However during automatic beam-out recovery the local controller will be able to transition between DISCHARGE 494

and THRUST CONTROL autonomously. 495

2. Electrical Compatibility 496

The primary purpose of Coupling Test is to verify electrical compatibility of the PPU, FCU and Thruster and 497

demonstrate that the Thruster and FCU perform nominally when driven by the PPU. The verification will also 498

extend to the PPU driving the Thuster in a redundant harness configuration, representing the redundant switching of 499

the PPU from the primary DANS terminals to the switched terminals through the Cross-Strap Splice Plate. The 500

compatibility tests are performed in discharge and at the four nominal thrust levels of 75mN, 100mN, 125mN and 501

145mN at two input conditions: nominal SEPS input and minimum SEPS input. The minimum SEPS input case 502

refers to the condition where the SEPS is only provided with the minimum FCU inlet pressure & minimum PPU 503

spacecraft power bus voltage (98V). 504

[11]: S/C

[12]: Auto

[13]: Auto, S/C

[14]: S/C, FDIR

[15]: S/C, FDIR

[16]: S/C, FDIR

[17]: S/C, FDIR

[18]: S/C, FDIR

[19]: S/C, FDIR

[21]: S/C

[22]: Auto, S/C, FDIR

HP:

S/C:

FDIR:

Auto:

State transition caused by high power

enable or disable signal

State transition commanded by spacecraft

State transition forced by FDIR

State transition controlled by normal or

recovery sequence operation

[01]: HP enable

[02]: S/C

[03]: S/C

[04]: S/C

[05]: S/C, FDIR

[06]: HP disable(1)

(1): Note that HP disable is

available in all states

[19]

STANDBY

CONFIGURE

SHORT CLEARMANUAL START UP

DISCHARGE

THRUST

CONTROL

OFF

[01][06]

[15]

[22]

[12][16]

[13][14]

[21]

[18]

[17]

STATE TRANSITIONS

STATE TRANSITIONS SOURCE DESCRIPTIONS

[04]

[03]

[05]

[02]

[11]

Figure 13. SEPS Modes/Transitions


October 6 – 10, 2013

11

Prior to the coupling tests, the thruster was operated with FGSE and Thruster driver EGSE rack to commission 505

the Thruster and generate a baseline performance 506

data set. The performance dataset was then used to 507

validate the later coupling tests and verify that the 508

thruster is performing within nominal parameters. 509

Table 1 shows the comparison of the electrical 510

data at 145mN thrust captured during the 511

shakedown test using EGSE/FGSE and (a) The 512

Nominal Harness configuration with both nominal 513

and minimum SEPS inputs and (b) The Redundant 514

Harness configuration with both nominal and 515

minimum SEPS inputs. 516

The data shows excellent agreement in engine 517

performance between the coupling tests and the 518

shakedown test, with the small voltage differences 519

being attributable to different harness 520

configurations. The results demonstrated the 521

electrical compatibility of the SEPS equipment. 522

523

3. Thrust Control/Ramp 524

When commanded by the OBC, the SEPS is capable of autonomously going to a desired thrust level in 525

predetermined steps. The PPU starts the cathode heaters, enables the flows to the three branches, initiates the 526

discharge, goes into thrust mode and ramps the thrust from 0 to 75mN in 16 steps and then ramps onwards up to the 527

desired thrust level at a rate of 0.5mN/s . 528

The SEPS coupling test was required to demonstrate this capability and to verify that all the ramp profiles were 529

achievable in terms of thruster electrical data and input power requirements. The minimum SEPS power demand, for 530

a given thrust level, usually occurs with the thruster in thermal equilibrium. The peak power conditions, usually 531

arise just when the thruster achieved the desired thrust level. The thruster initial conditions are important in 532

determining peak power demand; the colder the engine at the moment of discharge strike, the higher the peak power 533

demand in a ramp. Consequently, to determine the SEPS peak power demand, the engine was subjected to an 534

extended cold condition prior to discharge initiation. Moreover, this profile was carried out at minimum FCU input 535

pressure (2.904bar) and minimum Power bus input voltage (98V) which demonstrated the SEPS ability to achieve 536

the required thrust levels at the worst case PPU and thruster input conditions. 537

Figure 14 shows the timing information and electrical data for the representative thrust ramp to the maximum 538

SEPS capability of 145mN. The cathode and neutraliser 539

heaters were initiated first and when the heater voltage 540

reached a predetermined value, the neutraliser 541

heater/keeper supply is disabled and switched to apply 542

the output to the keeper. The neutraliser discharge is 543

struck and maintained at a constant current value of 544

4.2A. Similarly when the cathode heater supply reached 545

its predetermined value the anode power supply is 546

enabled with a constant current value of 5A to strike the 547

main discharge. Simultaneously the heater/magnet 548

supply is disabled and switched to apply output to the 549

magnet, which is enabled at a constant current value of 550

0.4A. After a period of discharge dwell, the beam supply 551

is enabled and the thruster is ramped up to 75mN in 16 552

steps. The PPU then proceeds to ramp up the thrust in 34 553

steps, using the flight representative 0.5mN/s criterion, 554

till the full thrust level of 145mN is reached. 555

Once at the desired 145mN beam current, the thrust 556

control algorithm uses the beam current as the primary feedback parameter. Any thrust level has an associated anode 557

current whereas the solenoid current is variable (between minimum and maximum limits) and is used to ‘trim’ the 558

discharge conditions and therefore the beam current. It can be noted that almost as soon as the desired anode and 559

145mN

EGSE FGSE

Nominal Harness Redundant Harness

100V/nom

FCU inlet

98V/min

FCU inlet

100V/nom

FCU inlet

98V/min

FCU inlet

IA 18.01 A 18.04 A 18.05 A 18.04 A 18.04 A

IM 0.85 A 0.86 A 0.89 A 0.88 A 0.89 A

IB 2.16 A 2.16 A 2.17 A 2.16 A 2.16 A

IAcc 13.97 mA 14.12 mA 13.85 mA 13.9 mA 14.02 mA

INK 4.2 A 4.19 A 4.19 A 4.19 A 4.19 A

VA 32.88 V 31.92 V 31.89 V 33.44 V 33.42 V

VA(S) 30.59 V 31.14 V 31.08 V 31.23 V 31.21 V

VA(corr) 29.41 V 29.69 V 29.65 V 29.81 V 29.78 V

VM 17.28 V 16.75 V 17.28 V 18.14 V 18.38 V

VB 1850 V 1845.43 V 1846 V 1845.9 V 1845.65 V

VACC -265 V 264.81 V 264.82 V 264.83 V 264.82 V

VNK 15.93 V 15.86 V 16.6 V 16.9 V 17.72 V

NRP -10.92 V -11.41 V -11.43 V -11.7 V -11.76 V

Table 1. SEPS 145mN Electrical Compatibility

Verification Data

-15

-10

-5

0

5

10

15

20

25

30

35

Vo

lta

ge

(

V)

0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

0.016

0.018

0.02

Acce

l C

urr

en

t (A

)

BI

MagCHI

NKNHI

AnI

AnVPPU

NRPPPU

AccI

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

Be

am

- M

ag

/CH

- N

K/N

H C

urr

en

t (A

)

08/05 12:28:48

08/05 12:36:00

08/05 12:43:12

08/05 12:50:24

08/05 12:57:360

2

4

6

8

10

12

14

16

18

20

An

od

e C

urr

en

t (A

)Representative Ramp and Discharge Dwell from Cold

Figure 14. Representative thrust ramp to 145mN


October 6 – 10, 2013

12

beam currents are achieved, the magnet current begins to ramp down as the engine warms up. This also has an 560

associated effect on the anode voltage, which also begins to drop as the engine heads towards thermal equilibrium. 561

562

Figure 15 displays the Power demand profile 563

during the representative thrust ramp obtained using a 564

scope. The figure contains 500sec of data, which 565

encompass the end of the discharge dwell period, the 566

16 initial steps and the flight representative ramp to 567

145mN. Critically it manages to capture the peak bus 568

current demand, with their location and magnitude 569

indicated on the figure. 570

571

The thrust control and ramp tests have verified that 572

the SEPS has successfully executed an autonomous 573

thrust ramp from a cold condition to 145mN thrust at 574

peak power demand and at the worst SEPS PPU and 575

thruster input conditions. 576

577

4. Thrust Quantization 578

The SEPS coupling test presented an 579

opportunity to measure the minimum thrust step 580

increment capability of the SEPS (referred to as 581

SEPS thrust quantization). The thrust steps were 582

inferred by measuring the beam current step 583

directly at the beam test port using an 584

oscilloscope to obtain the quantized waveforms 585

and then post processing the data. The thrust 586

increments were achieved by modulating the 587

Magnet Current by the smallest possible 588

increments (i.e. 1 hex = 3mA, 2 hex = 6mA and 3 589

hex = 9mA) at each of the nominal thrust levels: 590

75mN, 100mN, 125mN and 145mN. 591

Figure 16 shows the obtained 145mN Beam and Magnet 592

current steps and illustrates the use of the scope’s built in 593

persistence function to visually highlight and recover the 594

steps from the trace noise. Table 2 summarises the results for 595

the nominal thrust levels. 596

The results have shown that the SEPS is capable of 597

stepping the thrust in as little as 0.021 to 0.177mN steps 598

depending on the thrust level. 599

600

5. Beam-Out Recovery 601

In a beam-out, a piece of sputter or debris intercepts the 602

gap between the screen and accel grid, which causes a short. 603

The short circuit will cause transient currents in the beam 604

supply modules. This is the most taxing and demanding 605

condition for a gridded ion engine system. The SEPS must: 606

(1) Demonstrate a tolerance to a beam-out condition i.e. 607

electrical compatibility of the supplies with transients (2) 608

Include means for detection of the transient and a transient 609

counter (in case a high beam out frequency is a symptom of 610

another underlying problem) (3) Be able to autonomously recover from the beam event and return to its prior thrust 611

level within a set time. 612

Scope current and voltage probes were attached to critical lines into and out of the PPU to obtain high frequency 613

monitoring of the beam-out events and subsequent recovery. Figure 17 shows scope traces of such a beam-out and 614

Figure 15. Power bus data for representative thrust

ramp to 145mN

Figure 16. 145mN Thrust Quantization Persistence Trace

Table 2. Thrust Quantization results table


October 6 – 10, 2013

13

recovery case at 145mN .The figure shows two scope plots wherein scope1 is triggered by the collapse in Beam 615

voltage and scope2 is simultaneous triggered into acquisition by TTL signal from scope1. 616

617

In Fig. 17 the top scope plot, traces of 618

the NRP, Beam and DANS Bus voltages 619

and DANS Bus current are shown. In the 620

bottom scope plot, traces the Neutraliser 621

Keeper and Beam Supply Bus voltages and 622

Anode+Beam and Beam Supply Bus 623

currents are shown. 624

At the onset of the beam-out, the Beam 625

voltage collapses resulting in the expected 626

reductions in the NRP voltage and the 627

Beam Supply Bus current. The 628

Anode+Beam current is also seen to reduce 629

to the discharge only current of 18A 630

(discharge current at 145mN minus Beam 631

current). Approximately 1s later the Anode 632

supply is commanded to reduce to 5A and 633

the Beam Supply is re-enabled. Following 634

the establishment of the beam, the Anode + 635

Beam current begins to ramp back up to the 636

value for the original thrust point (with 637

both Power Bus currents following the 638

trend) and the beam-recovery sequence is 639

then complete. 640

A number of automatic beam-out 641

recoveries at 75, 100, 125 and 145mN have 642

been observed and recorded. At PPU nominal Bus input voltage levels every beam-out that has been experienced 643

(regardless of the thrust level) is automatically recovered. 644

6. Failure Detection Isolation and Recovery and Hardware Protections 645

A number of surveillances and hardware protections are built into the SEPS to protect the system and to improve 646

reliability. Surveillances resident in the PPU is active in all DANS states by default apart from CONFIGURE, 647

STANDBY and OFF (with some exceptions). Hardware protections are implemented as hardware trips and hence 648

are active in all DANS states apart from OFF. 649

The surveillances are designed to detect unusual SEPS behaviors’ and when one of these occurs either abort the 650

operation to prevent/minimise damage to the SEPS or initiate auto recovery to maintain safe operation. 651

All surveillances have been tested with the thruster in a representative environment. A 652

V. Future Tests 653

D. SEPS Coupling Test Part 2 654 As described previously, following on from the success of the SEPS EQM Coupling Test, there are further 655

planned tests to demonstrate the EQM equipment’s, one of these tests in a further coupling test, where the SEPS 656

hardware will be mounted inside of the vacuum facility to represent a complete single chain (i.e. no break in the 657

electrical harness to break the vacuum seal). 658

659

Figure 17. Beam out and recovery at 145mN (Scope data)


October 6 – 10, 2013

14

1. Test Objectives 660

To provide a platform for qualification of the SEPS system. The main purpose is to perform performance 661

analysis of coupled SEPS hardware. It 662

also demonstrates specific PPU-Thruster 663

interactions including the control 664

algorithms, beam-out recovery, FDIR etc. 665

and is a means of formal requirement 666

verification. 667

The main purpose of this test 668

campaign is to perform detailed 669

performance analysis of coupled SEPS 670

hardware. The SEPS coupling test would 671

also provide verification of the following: 672

Electrical compatibility of anode, 673

heater/magnet, accel, neutraliser 674

keeper/heater, beam; FCU electrical 675

interfaces; FCU control loops (algorithm); 676

SEPS command and control sequences 677

including thrust ramps; Electrical compatibility of supplies with transients (e.g. beam outs) including Beam-Out 678

Recovery; Look-up tables; Modes/Mode transitions; TM/TC; FDIR; PPU minimum inlet voltage; FCU flow 679

variation. 680

681

2. Test Set-Up 682

The test set-up has been designed so that the SEPS hardware will be located in the QinetiQ LEEP3 Vacuum 683

Chamber (schematic shown in Fig. 19 and CAD representation shown in Fig. 19). Careful integration activities have 684

been planned so that all of the electrical connections 685

to the PPU are made on a support trolley in a clean 686

tent outside of the vessel and the entire PPU is lifted 687

into the rear of the chamber where it will be mated 688

with the thruster and FCU. Due to the well-known 689

problem of ground testing contamination whilst firing 690

a thruster in a vacuum chamber (back sputter), the 691

PPU and FCU will be suitably shielded from the 692

target sputter, minimising performance issues or test 693

anomalies during the test campaign. The PPU will be 694

thermally controlled for both hot and cold cases by 695

the means of a synthetic oil chiller unit (HCU), this 696

will provide thermal control of the unit when it is at 697

full operational power and hibernation, when the unit 698

is off, all of which have been fully demonstrated in 699

previous test campaigns. 700

The test will reply solely upon PPU TM/TC to validate requirements, both in the ‘direct’ and ‘switched’, across 701

the thrust range and all of the operational modes, where applicable. 702

E. SEPS Flight Acceptance 703 Due to the nature in which the SEPS hardware is integrated onto the spacecraft by the prime, it is not possible to 704

place the entire MTM structure, with all of the SEPS equipment attached and perform an acceptance firing. 705

Therefore to overcome these issues, a test has been developed by QinetiQ that satisfies the architecture that 706

demonstrates that each FM SEPS chain and its switched configuration are fired together and performance data 707

recorded for such couplings. This entails a much simpler test set-up (although one’s with its challenges all the 708

same), and one that will be discussed in this paper. 709

710


The main objectives of this campaign is to perform and acceptance firing of the SEPS FM models (excluding 712

SEPH and SEPP, as these will have already been integrated onto the spacecraft structure and TPM’s), and verify the 713

program acceptance requirements. The test will be split into two distinct parts, which allows all four thrusters and 714

Figure 18. Schematic of SEPS Coupling Test Pt2 Set-Up

Figure 19. CAD Model of PPU/FCU inside of LEEP3


October 6 – 10, 2013

15

their associated FCU’s to be coupled together with the two PPU’s (located outside of the vacuum chamber to protect 715

the flight PPU’s from sputter contamination) and the associated ‘direct’ and ‘switched’ harnesses. A key feature of 716

this test campaign is the firing of two thrusters simultaneously (as performed in the development tests reported in 717

section VE3) where the main objective is to establish the effects of simultaneous operation and measure any thrust 718

vector change when the other thruster is in operation, thruster to thruster interactions and continuous operations 719

during beam events (all of which have seen to be negligible in similar previous tests). 720

721

2. Test Set-Up 722

Since the FM SEPHs are not employed the FM PPUs can be located outside of the vacuum chamber, connecting 723

to the thrusters via chamber feed-throughs. The cross-strap elements, FCU elements and flexible elements will be 724

fully flight representative (with exception of 725

routing shape). 726

The SEPS configurations tested are listed in 727

Fig. 20. A SEPS single branch 1 (comprised of 728

both PPUs, thruster#1&2, FCU#1&2 and 729

representative SEPH and SEPP elements) is 730

operated in 2 ‘Direct’ and 2 ‘Switched’ 731

configurations. 732

The SEPS is operated in a single twin thruster 733

operating configuration. Following the branch 734

testing the thrusters and FCUs are removed from 735

the facility. SEPS single branch 2 is then 736

configured (comprised of both PPUs, thrusters#3&4, 737

FCU#3&4 and representative SEPH and SEPP 738

elements) and is operated in 2 ‘Direct’ and 2 739

‘Switched’ configurations. 740

A schematic of the hardware configuration is 741

shown in Fig. 21, which represent the first 742

configuration of SEPS hardware. Although the 743

configuration is a lot more complex than the previous 744

(due to the two branches) a single chain, with ‘direct’ 745

and ‘switched’ has already been demonstrated 746

successfully, and local integration and operating 747

procedures have been developed to cope with such a complex test set-up. 748

One of the key pieces of MGSE and set-up for this test campaign is the twin thruster MGSE, which is a rigid 749

support structure that interfaces to LEEP3, and provides the means to mount two thrusters and their associated 750

Thruster Splice Plates (TSP). The MGSE will provide a representation of the exact distance from thruster centres 751

and the second thruster mounted at an angle to represent the worst case pointing mechanism angle with respect to its 752

adjacent thruster. It is intended to perform beam diagnostic measurements during this campaign using a beam probe 753

array aligned with one of the thrusters. The array probes 754

are fitted with collimators that reject the ions from the 755

second thruster, allowing determination of the single 756

thruster thrust vector during all firing configurations. The 757

MGSE that supports the two thrusters is shown in Fig. 22, 758

and support two thermally controlled thruster interface 759

plates, and accommodation of the thruster shields. These 760

discs have been employed to protect the thruster that is not 761

operating from back sputter from the thruster that is 762

operating (again this is a ground testing effect, and 763

precautions need to be in place to protect the flight 764

hardware). These cover discs will be driven by a vacuum 765

stepper motor. A detailed FMECA and interlock system 766

Config Ident Direct or

Switched PPU# DANS# SEPT# Seq.

Single SEPT#1&2 operations (LEEP2)

1 D PPU#1 DANS#1 SEPT#1 1

8 S PPU#2 DANS#4 SEPT#1 2



Simultaneous Twin SEPT Operations (LEEP2)

13

D PPU#1 DANS#1 SEPT#1

5

S PPU#2 DANS#3 SEPT#2

Single SEPT#3&4 operations (LEEP3)





Figure 20. SEPS Configurations for

Acceptance Tests

Figure 21. SEPS Schematic for Acceptance Tests

Figure 22. Twin thruster MGSE


October 6 – 10, 2013

16

has been employed to ensure the thrusters are protected in a failure or anomalous condition. During twin thruster 767

firing, both shields will be stowed into the open position. 768

3. Twin Thruster pre-development tests 769

A series of pre-development tests (reported in detail elsewhere7) have been carried out to give confidence in the 770

SEPS flight acceptance approach and to validate certain 771

concepts to be used in the flight SEPS. The tests demonstrated 772

and characterised twin simultaneous firing of T6 thrusters 773

each with its supporting EGSE and FGSE at QinetiQ’s 774

LEEP2 facility. The propulsion system was operated over a 775

wide range of operating conditions covering the mission 776

thrust range of 75mN (single engine) to 290mN (dual thrust). 777

The tests were conducted with a link between the two EGSE 778

racks, which can be opened and closed. This enabled three 779

distinct modes of operation at each thrust level: (1) Twin 780

thruster single Neutraliser (2) Twin thruster dual Neutralisers 781

with common neutraliser returns (3) Twin thruster dual 782

Neutralisers with isolated Neutraliser returns. No significant 783

variation in a single thruster performance was observed when 784

one or two thrusters were operating. It was also established 785

that the current transients during a beam-out in one engine did 786

not affect the running of the second. Figure 23 shows a photo 787

of taken of the thrusters in LEEP2 during twin thruster firing 788

at the maximum combined throttle point. 789

790

Beam probe plume measurements were 791

carried out on the BB thruster using 792

specially modified Faraday probes with 793

collimating tubes that reject ions from the 794

second engine. The tests demonstrated the 795

success of the collimator concept in selecting 796

ions emanating from only one engine and 797

showed the stability of the BB thrust vector 798

when the second engine is operated and 799

minimal interaction between the engine 800

plumes. Figure 24 shows Example results of 801

a 3D and a 2D contour plot of the probe 802

sweep when both the BB and TDA SEPTs 803

operated at 145mN. 804

F. SEPS EMC Test 805 The conventional approach for EMC tests of the PPU and FCU is unfortunately not directly applicable to the 806

SEPT because in order for it to be energised and thus radiating E and H fields, it must be under vacuum within a 807

substantial electric propulsion vacuum test facility. The same is also the case for the Radiated Susceptibility tests. 808

These vacuum chambers are invariably metal (usually stainless steel or aluminium) and hence the conventional 809

EMC test configurations cannot be applied. 810

To validate the EMC requirements of the SEPS hardware, the SEPT will be placed inside a large vacuum 811

chamber, with the centre section constructed from an RF transparent material (glass fibre), which is in turn placed 812

inside a large anechoic screened room. 813

The EMC test campaign will entail measurements of power quality, conducted and radiated emissions, and 814

radiated susceptibility of the BB SEPT driven by the EQM PPU. 815

816


The EMC tests to be performed shall address the following requirements: 818

Power Quality 819

In-rush Current 820

Figure 23. Twin thrusters operating at

maximum combined throttle point (290mN) in

LEEP2 vacuum facility

Figure 24. 3D (left) and 2D (right) Contour plots from twin

thruster testing at 290mN (145mN on each)


October 6 – 10, 2013

17

Voltage Transients 821

Conducted Emissions (CE) 822

Conducted Emissions on Primary Power Leads (frequency domain) 823

Conducted Emissions on Primary Power Leads (time domain) 824

Radiated Emissions, E-Field (REEF) 825

Radiated Emissions, H-Field (REHF) 826

Radiated Susceptibility, E-Field (RSEF) 827

Radiated Susceptibility, H-Field (RSHF) 828

The measurement of REEF, REHF, RSEF and RSHF require the Equipment Under Test (EUT) to be inside an 829

electromagnetically screened enclosure to reduce the background RF noise to minimum levels. The emissions and 830

susceptibility of the EUT can then be measured using a series of different antenna, receivers and transmitters 831

covering the frequency range of interest. 832

833

2. Test Set-Up 834

The SEPT EMC testing is conducted inside a modified Thermal Vacuum Chamber, located at the QinetiQ Space 835

Test facilities, Farnborough (illustrated in Fig. 25). The 836

existing vacuum chamber has been modified by inserting an 837

RF transparent section of GFRP ~1.6m diameter x 3.0m 838

long. The RF transparent section will be enclosed by a 839

screened enclosure lined with RF absorbing material. While 840

the design aims to meet Mil-Spec 461F by providing an 841

ambient RF noise level inside the enclosure that is better 842

than 6 dB below the RF noise level outside, it is recognised 843

that the intrusion of the vacuum chamber presents a 844

compromise over a more ideal construction. Consequently, 845

it will be necessary to perform a series of RF evaluation 846

tests in the enclosure to characterise its RF environment. 847

The results of such evaluation tests will be used in 848

conjunction with the SEPS EMC test results to ensure a 849

credible EMC assessment is made. 850

As in the previous Coupling Tests, the switchover between the long and short harness is achieved autonomously 851

by switching the DANS outputs to the representative 852

harnesses. All of the operating thruster EMC testing will 853

be conducted with the thruster inside the RF transparent 854

section of the vacuum chamber, shrouded by the screened 855

anechoic room. The PPU will be outside the screened 856

room for these tests. A top level configuration is presented 857

in Fig. 26 showing the hardware location for the thruster 858

and SEPS testing. This particular diagram shows an 859

antenna for E field radiated emissions / susceptibility. 860

However, the EGSE and PPU would be configured 861

similarly for all the SEPS and thruster tests emissions / 862

susceptibility. 863

864

865

VI. Conclusion 866

This paper has presented the BepiColombo SEPS coupling test that was successfully completed using QinetiQ 867

EP test facilities at Farnborough. The SEPS coupling test was a major risk retirement activity completed ahead of 868

the build of flight SEPS hardware. 869

The SEPS coupling test demonstrated the electrical compatibility and system interactions at qualification level. 870

In addition, the SEPS coupling test demonstrated the systems automatic recovery following a beam interruption 871

(these beam interruption events are enhanced due to ground testing effects). Finally the SEPS coupling test 872

demonstrated autonomous operation and control of the SEPS. 873

Figure 25. EMC Test Facility

Figure 26. EMC Test Set-Up


October 6 – 10, 2013

18

Acknowledgments 874

S. D. Clark thanks the BepiColombo teams within ESA, Astrium UK and Astrium Germany for their support to 875

date in the successful development of SEPS. In addition, S. D. Clark thanks EADS Astrium Crisa and Moog- 876

Bradford Engineering for their efforts in developing the PPU and FCU and towards the completion of the SEPS 877

Coupling Test. M. S. Hutchins thanks ESA’s ARTES 3-4 for its financial support for the development and 878

qualification of HPEPS. 879

References 880 1Novara, M, “The BepiColombo ESA Cornerstone Mission to Mercury”, IAF Paper IAF-01-Q.2.02, 2001. 881 2Wallace, N. C., “BepiColombo Technology Demonstration Activity (TDA) –Simultaneous Twin Thruster Firing Tests and 882

500 Hour Thermal Endurance Test Report”,QINETIQ/KI/SPACE/TR031835 (Issue 2.0), August 2004. 883 3Wallace, N. C., “BepiColombo Technology Demonstration Activity (TDA) – Electric Propulsion Technology Assessment 884

Study”, QINETIQ/KI/SPACE/TA030077 (Issue 1.0), February 2003. 885 4Wallace, N, and Fehringer, M, “The ESA GOCE mission and the T5 ion propulsion assembly”, IEPC Paper 09-269, 2009. 886 5Wallace, N., Saunders, C. and Fehringer, M. “The in-orbit performance and status of the GOCE ion propulsion assembly 887

(IPA)”, Space Propulsion, 2010. 888 6Kaufman, H R, “An ion rocket with an electron bombardment ion source”, NASA TN-585, 1961. 889 7I F M Ahmed Rudwan , N Wallace & S Clark. "Twin Engine Tests of the T6 Ion Engine for ESA's BepiColombo Mercury 890

Mission", IEPC-2011-125, 32nd International Electric Propulsion Conference, Wiesbaden, Germany, 2011 891

bepicolombo electric propulsion thruster and high power

Documents